Danny on /x/ vs. /h/

Danny Weir (mid May):
>However, contrast of /h/ and /x/ is more common.  You have both
>phonemes in Irish and Scots Gaelic, Welsh, Scots English, Dutch,
>German, Czech, Ukrainian (the /h/ is voiced in that language),
>Arabic, Hebrew, Aramaic, Somali, Azeri, Turkmen, Kazakh, Kyrgyz,
>Tatar-Bashkir, Uzbek (where /x/ is uvular), Uyghur, Kalmyk-Oirat,
>Persian, Kurdish, Pashto, Urdu, Kashmiri, Hmong and (I think)
>Vietnamese.  Of course many languages have one but not the other:
>English, Hungarian, Turkish, and Japanese only have /h/; Spanish,
>Russian, Khalkha Mongolian and Mandarin Chinese only have /x/.
I'm not questioning the accuracy of this, but do your sources for this
info give any minimal pairs for a /h/ vs. /x/ contrast? I am
particularly interested in cases where [h] and [x] constitute pretty
much the entire allophonic range of the two phonemes, or at least are
the primary, citation-form allophones.

I have seen /x/ attributed to Scots, and occasionally to English in
general as a marginal phoneme, but have not so far happened to find
any text able or willing to make the attribution with adequate
consideration or justification. (To me it seems unwarranted, but then
I'm not sure what the Pro case is.)

--And.